Car Damage Estimation System using Deep Convolutional Neural Network

Road accidents can happen at any time. Especially car accidents that have caused many injuries or deaths. When they happened, the vehicles are damaged and there are some cost to spend for both injured or vehicle restoration. One way to reduce such costs is to prepay for a car insurance. The insurance company issuing insurance certificates must assess the degree of damage to the car. The insurance company's damage assessment involves taking pictures of the accident car and send back to the insurance company or taking the car to the garage to assess the damage. These processes takes a long time to verify Hence, causing delays in repairing and approval of costs to customers. To determine and approve the cost of the customer's car insurance accurately and appropriately therefore the research proposal is to developed a program to assess the damage of cars caused by accidents using convolutional neural networks, one of the deep learning algorithms, to apply vehicle damage detection and vehicle damage assessment. that can be used on mobile phones By taking a picture of the car and sending it to the system to assess the level of damage. This makes it possible to recognize the damage point of the car in more detail. It can reduce the time for issuing insurance certificates and approve the insurance money to customers more quickly.